Our goal is to elucidate the principles that govern TGF signal transduction and mode of action. In the past 5 years we elucidated a SMAD transcriptional cycle including nuclear CDK8/9 and GSK3 kinases that drive Smad protein utilization and disposal in TGF and BMP pathways (Alarcn et al. Cell 2009; Gao et al. Mol. Cell 2009). In the 5-year cycle now ending we defined a molecular switch that couples the delivery of TGF signals with the turnover of the SMAD messengers (Aragn et al. Genes Dev. 2011; Aragn et al. Structure 2012). Some of the TGF-SMAD target genes mediate metastasis in colorectal and breast cancer (Caln et al Cancer Cell 2012; Stankic et al. Cell Reports 2013). Furthermore, we determined how signal-driven SMADs bypass repressive chromatin marks to regulate stem cell differentiation. A new arm of the SMAD pathway, involving Trim33 as a reader of repressive H3K9me3 histone marks, allows the canonical Smad arm to activate mesendoderm specification genes (Xi et al. Cell 2011). This progress now provides the opportunity to address two long-standing questions: How does the TGF-SMAD pathway regulate stem cell differentiation? And, how does this signaling mediate tumor suppression in premalignant cells? Each question is the subject of two Specific Aims. Our Aim 1 is to elucidate how activated SMADs target master differentiation genes in embryonic stem cells. We will define how activin/nodal-driven SMAD transcriptional complexes recognize specific regulatory elements in the mesendoderm specification gene goosecoid (Gsc), a model master differentiation gene. In related work, Aim 2 will address the mechanistic basis for the remarkable role of p53 as a requirement for SMAD action on Gsc. Also instigated by recent progress, our Aim 3 is to understand how SMAD-dependent cell differentiation underlies tumor suppression. We recently found that premalignant pancreatic progenitors respond to TGF with a lethal combination of EMT and epithelial differentiation programs. Related to this phenomenon, our Aim 4 is to define how RAS and AKT act as determinants of TGF-mediated tumor suppression. We will address how mutant KRas in pancreatic progenitors enables a pronounced Snail response to TGF leading to a lethal EMT, and how PI3K-Akt signaling inhibits TGF- mediated tumor suppression. With this set of interrelated aims we seek to provide new insights into the mechanisms that regulate stem cell differentiation and cancer cell elimination by the TGF-SMAD pathway. By illuminating the operating logic of these cell decisions we will provide a better understanding of TGF tumor suppression and potentially pave the way for the reactivation of these mechanisms for the treatment of cancer. Through this work we hope to forge innovative paths at the forefront of stem cell and tumor biology while retaining the long-term focus of CA34610 on the basic principles of TGF action.